| 04:08.47 | *** join/#asterisk-doc alerios (n=alerios@201.244.242.109) |
| 08:34.49 | *** join/#asterisk-doc ToTo (n=ToTo@host212-207.pool872.interbusiness.it) |
| 09:21.16 | *** join/#asterisk-doc ToTo (n=ToTo@host212-207.pool872.interbusiness.it) |
| 10:11.29 | *** join/#asterisk-doc MatsK (n=mats@141.221.181.62.in-addr.dgcsystems.net) |
| 11:36.52 | *** join/#asterisk-doc champster (n=asterisk@AH.tescogroup.com) |
| 11:36.59 | *** part/#asterisk-doc champster (n=asterisk@AH.tescogroup.com) |
| 13:22.48 | *** join/#asterisk-doc Alric (n=nbowyer@masq.hyperusa.com) |
| 14:02.32 | *** join/#asterisk-doc alerios (n=alerios@201.244.242.109) |
| 14:23.44 | *** join/#asterisk-doc jsmith (n=jsmith@64.72.210.10) |
| 14:23.47 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 14:30.56 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 14:55.23 | jsmith_ | Too quiet in here... |
| 14:55.49 | Alric | NOISE! |
| 14:57.32 | Alric | I broke my script. |
| 14:58.20 | Alric | The heck. |
| 14:59.50 | Alric | PHP seems to be defying the laws of PHP. |
| 15:01.41 | Alric | Oh well. Fixed it. |
| 15:08.41 | jsmith_ | PHP is like physics -- you only *think* you understand the underlying laws |
| 15:09.52 | Alric | haha |
| 15:11.58 | Alric | Damnit |
| 15:12.12 | Alric | I upgraded this box to 1.2.7 zaptel and 1.2.10 asterisk |
| 15:12.20 | Alric | and the whole box has locked up, twice in 2 days.w |
| 15:12.21 | Alric | days* |
| 15:14.16 | Alric | And died without giving any log indication of what might be going on. |
| 15:16.02 | Alric | Well, thats not totally right. The box goes into some sort of deadlock. It will respond to pings, but nothing else. |
| 15:19.07 | jsmith_ | Really? |
| 15:19.12 | jsmith_ | man, it's been good to me. |
| 15:20.02 | Alric | Jul 27 10:04:06 WARNING[1076] channel.c: Avoided initial deadlock for '0x824c708', 10 retries! |
| 15:20.06 | Alric | I get bizillions of those. |
| 15:20.18 | Alric | One practically for each call setup / teardown. Didn't happen when the box was on 1.0.x |
| 15:20.32 | jsmith_ | Yuck, that's bad. |
| 15:20.35 | jsmith_ | What are you doing? |
| 15:20.42 | Alric | Nothing! |
| 15:20.51 | jsmith_ | AgentCallbackLogin()? |
| 15:20.54 | Alric | Its a PRI to a box and then Cisco 7960s off that |
| 15:20.55 | Alric | Nope |
| 15:21.06 | Alric | SIP interfaces straight into the Queue |
| 15:21.35 | jsmith_ | Ah, but there's a Queue involved! |
| 15:21.41 | Alric | Yeah? |
| 15:22.04 | Alric | But it happens with DID calls as well, things that totally bypass the Queue. |
| 15:22.07 | Alric | Jul 27 09:49:27 NOTICE[1121] chan_zap.c: PRI got event: HDLC Bad FCS (8) on Primary D-channel of span 1 |
| 15:22.13 | Alric | Get those too. I blame the megaraid card. |
| 15:22.24 | jsmith_ | I blame your PRI. |
| 15:22.30 | jsmith_ | You've got framing errors on the line. |
| 15:22.52 | Alric | So thats not the megaraid card holding the interrupt too long? |
| 15:23.11 | jsmith_ | Doubt it. |
| 15:23.26 | jsmith_ | You could put a loopback plug in the T1 card and run a pattern test. |
| 15:23.29 | Alric | Framing errors... |
| 15:23.36 | Alric | Would those show up on a QRSS test? |
| 15:23.47 | Alric | or all 1s, all 0s, 1 in 8, all that jazz? |
| 15:24.01 | jsmith_ | Yeah, they should. |
| 15:24.22 | jsmith_ | What usually happens is that they test the line when you first turn up your circuit, but then the errors show up a week later or so... |
| 15:24.39 | jsmith_ | (once they start punching down stuff close to your circuits) |
| 15:25.02 | Alric | Hmm. |
| 15:25.16 | Alric | But this PRI thing wouldn't be causing an entirely deadlocked box, would it? |
| 15:28.16 | Alric | <PROTECTED> |
| 15:28.16 | Alric | Jul 27 09:58:00 WARNING[1072]: channel.c:787 channel_find_locked: Avoided initial deadlock for '0x820a1a0', 10 retries! |
| 15:28.20 | Alric | Whee. |
| 16:16.07 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 16:18.48 | Alric | As long as it doesn't stay up for more than 24 hours, it'll probably be fine. |
| 17:02.41 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 18:29.15 | *** join/#asterisk-doc jsmith (n=jsmith@64.72.210.10) |
| 18:29.15 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 18:53.39 | jsmith | OK, so queues on 1.2.10 just plain suck |
| 18:55.11 | Alric | Wonder why. |
| 18:55.23 | Alric | And why did you arrive at this conclusion so suddenly? |
| 18:58.55 | *** join/#asterisk-doc jsmith (n=jsmith@64.72.210.10) |
| 18:58.55 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 19:06.39 | jsmith | Alric: Because of my experiences today |
| 19:06.54 | Alric | I wonder why they suddenly got bad. |
| 19:10.56 | *** join/#asterisk-doc Alric (n=nbowyer@masq.hyperusa.com) |
| 19:10.58 | Alric | Ugh. |
| 19:11.01 | Alric | Colloquy just died |
| 19:15.20 | *** join/#asterisk-doc g-bags (n=gra@86.42.2.209) |
| 19:15.44 | *** part/#asterisk-doc g-bags (n=gra@86.42.2.209) |
| 19:19.31 | Alric | What happened today? |
| 19:54.54 | jsmith | Queues broke, obviously |
| 19:55.43 | Alric | Hmmm |
| 19:55.49 | Alric | I wonder what changed. |
| 19:56.03 | Alric | That would poison queues suddenly :/ |
| 19:57.02 | jsmith | Could be that the T1 provider was making changes while I was doing my test. |
| 19:57.09 | jsmith | Or, it could be that 1.2.10 sucks |
| 19:57.19 | jsmith | Not really sure yet... doing another test this afternoon. |
| 19:57.53 | Alric | This 8-5 job is wrecking my response ability :( |
| 20:03.13 | Alric | Hmmm. 60ms ping times. Across a LAN. |
| 20:04.22 | Alric | Meh. My ethereal script I wrote? pcntl_fork doesn't work on older versions of PHP. Like the latest that comes with RH9 :( |
| 20:05.11 | Alric | 4.2.2. Wow. |
| 20:07.39 | Alric | Do you let the yum daemon run on your servers? |
| 20:08.15 | jsmith | Nope... |
| 20:08.20 | jsmith | But then again, I don't run RH9 either. |
| 20:08.43 | Alric | I meant on your CentOS boxes. |
| 20:09.10 | jsmith | Yum daemon... nope. |
| 20:09.14 | jsmith | I run yum on my terms |
| 20:13.41 | Alric | This is so strange. How in the world does faxing work across this wireless link, but phones themselves seem to have trouble. |
| 20:14.27 | Alric | You'd think that if phones were having trouble faxing would be impossible. |
| 20:14.30 | Alric | (FoIP) |
| 20:15.19 | *** join/#asterisk-doc blitzrage (n=blitzrag@asterisk/documenteur-extraordinaire/blitzrage) |
| 20:15.19 | *** mode/#asterisk-doc [+o blitzrage] by ChanServ |
| 20:16.11 | Alric | Anyone know of an affordable 24-port switch that does QoS on a traffic-like basis, not a port-like basis? As in strictly by the ports on the switch? |
| 20:17.20 | blitzrage | sounds like a layer 3 switch to me |
| 20:17.32 | Alric | Right, so !affordable :/ |
| 20:17.41 | blitzrage | exactly |
| 20:17.42 | Alric | This Linksys doesn't appear to do it. |
| 20:17.45 | blitzrage | LOL |
| 20:17.48 | blitzrage | of course it doesn't |
| 20:17.54 | Alric | It says QoS :P |
| 20:17.59 | Alric | But its per switch port. |
| 20:19.13 | blitzrage | right |
| 20:19.32 | blitzrage | you want it to do it on the backplane -- sounds like a router, not a switch, to me |
| 20:19.56 | blitzrage | QoS is done at the IP level usually, not ethernet |
| 20:20.19 | blitzrage | there's no way for the switch to analyze the traffic unless it can look at the layer 3 packets, not the frames |
| 20:23.35 | Alric | ugh. |
| 20:23.41 | Alric | I think I'm going to be digging a trench. |
| 20:23.55 | Alric | and putting ethernet cabling in the ground. Ugh! |
| 20:24.57 | blitzrage | fun stuff |
| 20:25.01 | Alric | Yeah |
| 20:25.06 | Alric | Hrm. |
| 20:25.09 | Alric | Probably over 1500 feet too. |
| 20:25.22 | jsmith | Alric: Fiber + media converters |
| 20:25.29 | blitzrage | jsmith: it'll drop in 10 mins I imagine |
| 20:25.31 | Alric | How much does that run? |
| 20:25.47 | jsmith | Alric: It's not as bad as you'd think... |
| 20:25.55 | Alric | Whats that mean? :) |
| 20:26.18 | Alric | This client is a factory. |
| 20:26.36 | Alric | They linked their computers behind their phones, and their cabling is a mess! |
| 20:26.51 | jsmith | Alric: The media converters are $90 or so a piece, and the fiber isn't bad either... the hardest part is if you have to terminate your own fiber... |
| 20:27.02 | jsmith | Alric: Hopefully, you can pay someone to come run it for you. |
| 20:27.28 | Alric | I wonder how much a 2k ft run would cost. |
| 20:27.36 | Alric | Thats just an estimation. |
| 20:28.38 | blitzrage | so get multimode and not singlemode |
| 20:28.38 | blitzrage | $1,000,000 |
| 20:28.51 | Alric | heh |
| 20:29.00 | blitzrage | this is rediculous... |
| 20:29.05 | blitzrage | my Internet is sooooooo unstable |
| 20:29.08 | Alric | if it has more than 3 zeros, there's NWIH :P |
| 20:29.11 | blitzrage | how the heck am I supposed to get any work done? |
| 20:39.31 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 20:42.47 | Alric | You still in Canada Leif? |
| 20:42.55 | Alric | Whats this about internet issues? What changed? |
| 20:55.41 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 20:55.55 | Alric | Looks like jsmith_ is having net access issues too? :/ |
| 20:56.41 | jsmith_ | No, I just keep suspending this laptop to conserve battery |
| 20:59.16 | Alric | ah |
| 21:07.41 | jsmith_ | I really need to setup my IRC proxy... |
| 21:12.34 | jsmith_ | That is, after I setup my new colo server |
| 21:12.41 | jsmith_ | That is, when I get some free time. |
| 21:12.46 | jsmith_ | (Which will be never) |
| 21:15.44 | Alric | Going to die with a Todo list the size of Mt. Rainer? |
| 21:17.02 | jsmith_ | The plan is to finish it up two or three days after I die... |
| 21:17.51 | jsmith_ | It's too bad I'm not Bhuddist... then I'd have several more lifetimes to learn it all.. |
| 21:19.20 | Alric | Who wants to come back as a chopped steak though? |
| 22:24.15 | *** join/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |
| 23:55.06 | *** part/#asterisk-doc jsmith_ (n=jsmith@64.72.210.10) |